The gates at American prisons can seem like revolving doors. People come in, do their time, and—within 3 years—half are back behind bars, according to U.S. Department of Justice statistics. Now, a scientist says he has nailed down one potential risk factor. An intriguing natural experiment that followed ex-cons displaced by Hurricane Katrina suggests that when former prisoners wind up moving to the same neighborhood, they are more likely to return to a life of crime.
The idea that neighborhoods shape behavior is not new. But this is the first time researchers have tested links between reincarceration and the presence of other former inmates in a large experiment, says David Kirk, a sociologist at the University of Oxford in the United Kingdom and the sole author of the new study. "We can’t forget about where people live, and neighborhood context influences behavior.”
Kirk was following the fate of parolees leaving Louisiana prisons after Hurricane Katrina when he noticed something unusual. Those parolees who moved away from their original New Orleans neighborhoods were less likely to land back in jail than those who returned. He theorized that higher concentrations of former prisoners in the old neighborhoods might have something to do with the difference.
It’s almost impossible to tease apart which factors drive something as complicated as crime. But in this case, the Katrina disaster offered an unplanned, natural experiment. After the flood, fewer newly released prisoners were moving to New Orleans. Instead, they were settling farther afield, in places like Lafayette and Baton Rouge. Kirk used these post-Katrina shifts—driven by an outside natural disaster rather than internal forces such as rising poverty or gentrification—to see whether changes in the concentration of people just out of prison affected reincarceration rates.
To do so, he compared reincarceration rates with the concentration of new parolees for 493 Louisiana communities broken down by ZIP code. He divided the communities into two groups—those that experienced no change in the concentration of parolees and those that experienced a change between late 2005 and late 2007. In total, he tracked more than 5000 parolees: 2859 released from September to December of 2005, right after Katrina, and 2555 people who left prison during the same period in 2006.
In ZIP codes where the concentration of newly released prisoners went up, Kirk found that people from the 2006 “class” were more likely to wind up in prison again. Where the density of ex-prisoners went down, so did the reincarceration rates. The difference was significant: For every new ex-con who moved into a neighborhood of 1000 people, the reincarceration rate for other new parolees living there went up by 11%, Kirk reports online this month in the Proceedings of the National Academy of Sciences. So if an ex-con’s average chance of returning to prison after just 1 year was 22%—as it was in 2006—an additional new parolee in the neighborhood boosted that chance to nearly 25%.
The numbers climb for each new parolee added. In some of the most affected neighborhoods—where five of every thousand residents were recent parolees—nearly 35% were back behind bars within a year of getting out. Because neighborhood problems such as chronic poverty can play a role in crime rates, Kirk used statistical analysis to try to filter out those effects. The pattern still held true.
What the study doesn’t say is why. Kirk wonders whether a concentration of former inmates can have a corrosive influence on a community’s respect for the legal system, setting the stage for more crime. It’s also possible that having more people with criminal backgrounds nearby can lead to more opportunities to commit crimes, he says.
The findings underscore the way legal policies—such as the growth of the U.S. prison population from the 1980s—can have a ripple effect in communities, says Nancy La Vigne, director of the Justice Policy Center at the Washington, D.C.–based Urban Institute. But the impact of more ex-prisoners in a neighborhood is hard to disentangle from other factors, she says. For instance, some neighborhoods get more attention from police, and that could skew the findings. “If you’re returning to a community that already has intensive police presence, the odds of being arrested are a lot higher regardless of whether your reoffending rate is actually higher,” she says.
Todd Clear, a criminologist and provost of Rutgers University, Newark, in New Jersey, says Kirk’s findings fit with his observations of the toll that high rates of imprisonment can take on neighborhoods. In Tallahassee, Florida, in the late 1990s, he found that when the number of people being imprisoned reached high levels, a neighborhood’s social institutions became destabilized and crime increased. Likewise, when the number of people returning from prison went up, so did crime. He has seen similar patterns in more recent work in Boston. “Those neighborhoods become places that are ‘prisonized,’ ” Clear says.
If the concentration effect is real, directly addressing where people move when they leave prison isn’t simple. One option is to make more housing available to inmates over a wider area, rather than in pockets where housing is cheap and former prisoners are already concentrated, Kirk says. Many states also require people to return to the county where they were convicted, helping steer people toward certain neighborhoods.
La Vigne says these results shouldn’t overshadow work that shows family ties can help people succeed after leaving prison—often in the neighborhoods where they came from. She followed 1500 men as they left prison and found that family support played an important part in the chance of success.
She says another way to reduce the number of people coming out of prison into certain neighborhoods is to imprison fewer people to start with, an idea that has gained traction in recent years among politicians and policy analysts.